a

r
X
i
v
:
h
e
p
-
t
h
/
0
0
0
2
1
8
6

v
2



1
6

M
a
r

2
0
0
0
hep-th/0002186
PUPT-1920
IASSNS-HEP-00/13
Comments on Noncommutative Perturbative Dynamics
Mark Van Raamsdonk
1
Department of Physics, Princeton University
Princeton, NJ 08544, USA
and
Nathan Seiberg
2
School of Natural Sciences, Institute for Advanced Study
Olden Lane, Princeton, NJ 08540, USA
Abstract
We analyze further the IR singularities that appear in noncommutative field theories on
R
d
. We argue that all IR singularities in nonplanar one loop diagrams may be interpreted
as arising from the tree level exchanges of new light degrees of freedom, one coupling to
each relevant operator. These exchanges are reminiscent of closed string exchanges in
the double twist diagrams in open string theory. Some of these degrees of freedom are
required to have propagators that are inverse linear or logarithmic. We suggest that these
can be interpreted as free propagators in one or two extra dimensions respectively. We
also calculate some of the IR singular terms appearing at two loops in noncommutative
scalar field theories and find a complicated momentum dependence which is more difficult
to interpret.
February 2000
1
mav@princeton.edu
2
seiberg@sns.ias.edu
1. Introduction
In this note we continue the analysis of the perturbation expansion of field theories
on noncommutative R
d
[1-20]. We consider theories in various dimensions defined by an
action
S =

d
d
x tr

1
2
(∂φ)
2
+
1
2
m
2
φ
2
+
¸
n
λ
n
n
. .. .
φ ⋆ φ ⋆ · · · φ

, (1.1)
where ⋆ is the noncommutative, associative star product defined by
f ⋆ g(x) = e
i
2
Θ
µν

x
µ

y
ν
f(x)g(y)|
y=x
, (1.2)
and Θ is a constant anticommuting noncommutativity matrix,
[x
µ
, x
ν
] = iΘ
µν
. (1.3)
In [13], perturbative properties of these noncommutative scalar field theories were
investigated through the explicit calculation of correlation functions. The ⋆-product form
of the interactions leads to a momentum dependent phase associated with each vertex of
a Feynman diagram. This phase is sensitive to the order of lines entering the vertex, so
different orderings lead to diagrams with very different behavior. As was first demonstrated
by Filk [1], planar diagrams (with no crossings of lines) differ from the corresponding
diagrams in the commutative theory only by external momentum dependent phase factors.
These graphs lead to single trace terms like (1.1) in the effective action, including divergent
terms which renormalize the bare action. The Feynman integrals for these graphs are the
same as in the commutative case, resulting in the usual UV divergences which may be
dealt with in the usual way by introducing counterterms.
Nonplanar diagrams contain internal momentum dependent phase factors associated
with each crossing of lines in the graph. The oscillations of these phases serve to lessen
any divergence, and may render an otherwise divergent graph finite, providing an effective
cutoff Λ
eff
=
1

θ
in cases when internal lines cross (θ is a typical eigenvalue of Θ
µν
) or
Λ
eff
=
1

−p
µ

2
)
µν
p
ν

1

p◦p
in cases where an external line with momentum p crosses an
internal line. In the latter case, we see that the original UV divergence is replaced with an
IR singularity, since taking p →0 results in Λ
eff
→∞.
This striking occurrence of IR singularities in massive theories suggests the presence
of new light degrees of freedom. Indeed, an analysis of the one loop corrected propagator
of φ reveals that in addition to the original pole at p
2
≈ −m
2
, there is a new pole at
1
p
2
= O(g
2
). This new pole can be understood as arising from the high momentum modes
of φ running in a loop. If we try to use a Wilsonian effective action with a fixed cutoff Λ,
these modes are absent, and indeed, we find that the p →0 limit is not singular, since the
effective cutoff Λ
eff
is replaced by the cutoff Λ when p ◦ p < 1/Λ
2
. Thus the p → 0 and
Λ → ∞ limits do not commute. In order to write a Wilsonian effective action that does
correctly describe the low momentum behavior of the theory, it is necessary to introduce
new fields into the action which represent the light degrees of freedom. In [13], it was shown
that the quadratic IR divergences in the two point functions of φ
4
in four dimensions or
φ
3
in six dimensions can be reproduced by adding a field χ
0
with action of the form
S
χ
0
=

d
d
x gχ
0
tr(φ) +
1
2
∂χ
0
◦ ∂χ
0
+
1
2
Λ
2
(∂ ◦ ∂χ
0
)
2
. (1.4)
With this action, the quadratic IR singularity in the two point function of φ is reproduced
by a diagram in which φ turns into χ
0
and back into φ.
It should be stressed that in Lorentzian signature spacetime with Θ
0i
= 0 the field
χ
0
is not dynamical. It is a Lagrange multiplier [13]. Yet, it does lead to long range
correlations. Even though it is not a propagating field in this case, we will loosely refer to
it as a particle.
These effects are very reminiscent of channel duality in string theory [13]. There, high
momentum open strings running in a loop have a dual interpretation as the exchange of a
light closed string. By this analogy, we may associate the field φ with the modes of open
strings, while χ
0
describes a closed string mode. We thus see that noncommutative field
theories are interesting toy models of open string theories. Other evidence to the stringy
nature of these theories is their T-duality behavior when they are compactified on tori and
their large Θ behavior [13].
Clearly, the appearance of these closed string modes is a generic phenomenon occurring
whenever the commutative theory exhibits UV divergences. It is surprising because the
zero slope limit of [21] is supposed to decouple all the higher open string modes of the string
as well as the closed string modes. In hindsight this phenomenon is perhaps somewhat less
surprising. The fields living on a brane are modes of open strings. The parameters in this
theory are the zero momentum modes of the closed string background in which the brane
is embedded. If the theory on the brane is not conformal, the renormalization group in
the theory on the brane changes the values of these parameters. Therefore, it is typical
for the zero momentum modes of the closed strings not to decouple. We now see that in
the noncommutative theories, the nonzero modes also fail to decouple.
2
Given this understanding of the quadratic IR singularities as poles of a light particle,
it is interesting to ask whether we can find a similar interpretation for the inverse linear and
logarithmic IR singularities that appear. These occur wherever linear or logarithmic UV
divergences appear in the commutative theories, including two point functions and higher
point interactions. It is the goal of this paper to provide some further understanding of
these logarithmic and inverse linear IR singularities.
In the next section, we determine the complete set of IR singularities in the low energy
effective action that arise from one loop graphs in various scalar theories. We show that
they may be reproduced by including a set of χ fields coupling to each relevant operator
of the theory, if we allow some of the fields χ to have propagators which behave like
ln(1/p ◦ p) or (p ◦ p)
−1/2
. In section 3, we point out that these propagators arise naturally
if the χ fields are actually free particles in extra dimensions, coupling to φs that live on a
brane of codimension two for logarithmic singularities or codimension one for (p ◦ p)
−1/2
singularities. This is natural given the analogy with string theory, since we associate the
φ particles with open string modes which live on a brane, while the χs are closed strings
which should be free to propagate in the bulk. In section 4, we consider higher loop
graphs in scalar field theory, and find IR singularities with more complicated momentum
dependence that are more difficult to interpret.
2. Low energy one loop effective action
In this section, we write down the complete set of IR singular terms in the one loop
1PI effective actions of various scalar theories. We will take φ to be an N × N matrix
since it is useful to see how the indices are contracted in the various terms. In particular,
it turns out that all IR singular terms take the form
tr(O
1
(p))tr(O
2
(−p))f(p) , (2.1)
where the Os are operators built out of φ, and f(p) diverges quadratically, linearly, or
logarithmically as p →0. We show that any such term may be understood as arising from
the exchange of a single scalar particle which couples separately to tr(O
1
) and tr(O
2
) and
which has a propagator f(p).
3
2.1. φ
3
theory in d = 4
We begin by considering the simple example of φ
3
theory in four dimensions
S =

d
4
x tr

1
2
(∂φ)
2
+
1
2
m
2
φ
2
+
g
3!
φ ⋆ φ ⋆ φ

. (2.2)
The commutative theory is superrenormalizable, and the only UV divergences are loga-
rithmic divergences in the one loop contributions to the 1PI effective action. These come
from the planar and nonplanar diagrams shown in figure 1 which contribute respectively
to tr(φ
2
) and tr(φ)tr(φ) terms in the effective action.
k
p
k
p
Fig. 1: Planar and nonplanar contributions to the one loop quadratic
effective action in φ
3
theory in four dimensions.
In the noncommutative theory, the planar diagram is unchanged, while the nonplanar
diagram becomes finite, cutoff by Λ
2
eff
=
1
p◦p
. Combining the two contributions, we find
that the one loop quadratic effective action (at finite cutoff) is
S
eff
= (2π)
4

d
4
p
N
2
tr(φ(p)φ(−p))(p
2
+ M
2
)

1
2
tr(φ(p))tr(φ(−p))
g
2
64π
2
ln

1
M
2
(p ◦ p + 1/Λ
2
)

+ ... ,
(2.3)
where M is the planar renormalized mass, corrected at one loop by the planar diagram in
figure 1 plus a counterterm graph.
The second term in (2.3) arises from the nonplanar diagram and contains a logarithmic
IR singularity for Λ = ∞ but not at finite cutoff. Thus, as in [13], the Λ →∞ and p →0
limits do not commute. In order to reproduce the correct low momentum behavior in a
Wilsonian action, we must introduce a new field.
As for the case of theories with quadratic divergences considered in [13], we introduce
a new field χ which couples to tr(φ)

d
d
x gχ(x)tr(φ(x)) . (2.4)
4
Now, suppose that χ has a propagator given by
χ(p)χ(−p) = f(p) . (2.5)
Then upon integrating out χ, the quadratic effective action for φ receives a contribution
S = (2π)
d

d
d
p
1
2
tr(φ(p))tr(φ(−p))

−g
2
f(p)

. (2.6)
Thus, we see that the logarithmic singularity in (2.3) may be reproduced by including a
coupling (2.4) in the Wilsonian effective action, if χ has a momentum space propagator
f(p) =
1
64π
2
ln

p ◦ p + 1/Λ
2
p ◦ p

. (2.7)
The numerator in the logarithm has been chosen to cancel the incorrectly cutoff logarithm
coming from the nonplanar φ loop in the cutoff theory.
In this theory, there are no additional singularities at higher loops, so this single new
field is enough to reproduce all IR singularities. In section 3, we will give a possible inter-
pretation of this logarithmic propagator, but first we turn to more complicated examples.
2.2. φ
4
theory in d = 4
As a second example, we consider φ
4
theory in four dimensions
S =

d
4
x tr

1
2
(∂φ)
2
+
1
2
m
2
φ
2
+
g
2
4!
φ ⋆ φ ⋆ φ ⋆ φ

. (2.8)
The effective action for the case where φ is not a matrix (N = 1) was computed in [13]
(for low momenta),
S
eff
= (2π)
4

d
4
p
1
2
φ(p)φ(−p)

p
2
+ M
2
+
g
2
96π
2
(p ◦ p +
1
Λ
2
)

g
2
M
2
96π
2
ln

1
M
2
(p ◦ p +
1
Λ
2
)

+ ...

+ (2π)
4

d
4
p
i
1
4!
φ(p
1
)φ(p
2
)φ(p
3
)φ(p
4
)δ(
¸
p
i
)

g
2

g
4
3 · 2
5
π
2
¸
i
ln

1
M
2
(p
i
◦ p
i
+
1
Λ
2
)


g
4
3 · 2
6
π
2
¸
i<j
ln

1
M
2
((p
i
+ p
j
) ◦ (p
i
+ p
j
) +
1
Λ
2
)

+ ...

.
(2.9)
5
In this case, the quadratic effective action has both quadratic and logarithmic IR sin-
gularities for Λ = ∞, while the quartic term has two types of logarithmic singularities. In
[13], it was shown that the
1
p◦p
term in the Λ →∞quadratic effective action is reproduced
by a Wilsonian effective action which includes a χ
0
field coupling to φ with action of the
form (1.4).
We now focus on the terms containing logarithmic singularities. It is illustrative to
generalize to the case of arbitrary N and write them as

d
4
xd
4
y

−g
2
M
2
tr(φ(x))tr(φ(y)) −
g
4
3
tr(φ(x))tr(φ
3
(y)) −
g
4
4
tr(φ
2
(x))tr(φ
2
(y))

·
1
3 · 2
6
π
2

d
4
p
(2π)
4
e
ip·(x−y)
ln

1
M
2
(p ◦ p +
1
Λ
2
)

=

d
4
xd
4
y
¸
g
m+n
M
4−m−n
γ
mn
tr(φ
m
(x))tr(φ
n
(y))∆(x −y) ,
(2.10)
where
∆(x −y) =

d
4
p
(2π)
4
e
ip·(x−y)
ln

1
M
2
(p ◦ p +
1
Λ
2
)

, (2.11)
and γ
mn
are numerical constants that may be read off from (2.10). In this form it is
clear that the complete set of logarithmic IR singularities in the Λ →∞ one loop effective
action may be reproduced with a finite cutoff Wilsonian action by including χ fields with
couplings

d
4
x
3
¸
n=1
g
n
M
2−n
χ
n
(x)tr(φ
n
(x)) (2.12)
and logarithmic propagators
χ
m
(p)χ
n
(−p) = −2γ
mn
ln(
p ◦ p +
1
Λ
2
p ◦ p
) . (2.13)
In this way, the Λ = ∞ IR singularity in each term of (2.10) is reproduced at finite cutoff
by the exchange of a single scalar particle, as shown in figure 2.
tr( )tr( ) φ φ
tr( )tr( ) φ φ
3
tr( )tr( ) φ φ
2 2
Fig. 2: Examples of nonplanar diagrams in φ
4
in d = 4 contributing IR
singularities for Λ →∞ and χ exchange diagrams that reproduce the singu-
larities at finite cutoff.
6
2.3. φ
3
theory in d = 6
As a final example, we consider φ
3
theory in six dimensions,
S =

d
6
x tr

1
2
(∂φ)
2
+
1
2
m
2
φ
2
+
g
3!
φ ⋆ φ ⋆ φ

. (2.14)
Here, the one loop effective action (for low momenta) was computed in [13] for N = 1,
S
eff
= (2π)
6

d
6
p
1
2
φ(p)φ(−p)

p
2
+ M
2

g
2
2
8
π
3
(p ◦ p +
1
Λ
2
)
+
g
2
3 · 2
9
π
3
(p
2
+ 6M
2
) ln(
1
M
2
(p ◦ p +
1
Λ
2
)
) + ...

+ (2π)
6

d
6
p
i
1
3!
φ(p
1
)φ(p
2
)φ(p
3
)δ(
¸
p
i
)

g +
g
3
2
9
π
3
¸
i
ln

1
M
2
(p
i
◦ p
i
+
1
Λ
2
)

¸
.
(2.15)
The IR singularities in this action at Λ = ∞ are similar to those of the φ
4
theory, but now
we have a p
2
ln(
1
M
2
p◦p
) term in the quadratic effective action. We may rewrite the terms
with logarithmic singularities for arbitrary N as

d
6
xd
6
y

gM tr(φ(x))

gMtr(φ(y)) −
g
6M
tr(∂
2
φ(y)) +
g
2
2M
tr(φ
2
(y))

·
1
512π
3

d
6
p
(2π)
6
e
ip·(x−y)
ln

1
M
2
p ◦ p

.
(2.16)
These may be reproduced by introducing χ fields with couplings

d
6
xgM χ
1
(x) tr(φ(x)) +χ
2
(x)

gMtr(φ(x)) +
g
2
2M
tr(φ
2
(x)) −
g
6M
tr(∂
2
φ(x))

(2.17)
and propagators
χ
1
(p)χ
2
(−p) = −
1
512π
3
ln

p ◦ p + 1/Λ
2
p ◦ p

χ
1
(p)χ
1
(−p) = χ
2
(p)χ
2
(−p) = 0.
(2.18)
By a linear redefinition of the χs, we may simplify the couplings to

d
6
x gM ˆ χ
1
(x)tr(φ(x)) + ˆ χ
2
(x)

g
2
2M
tr(φ
2
(x)) −
g
6M
tr(∂
2
φ)

, (2.19)
where ˆ χ
1
= χ
1
+ χ
2
and ˆ χ
2
= χ
2
.
7
2.4. General procedure
The three examples we have considered lead us to a general procedure for introducing
χ fields in the Wilsonian effective action in order to reproduce logarithmic singularities in
the one loop effective action. For a general scalar theory, logarithmic IR singular terms in
the effective action arising from one loop non-planar diagrams take the form
¸
m,n

d
4
xd
4
y
1
2
O
m
(φ(x))O
n
(φ(y))γ
mn

d
d
p
(2π)
d
e
ip·(x−y)
ln

1
m
2
(p ◦ p +
1
Λ
2
)

. (2.20)
Here {O
n
(φ(x))} is some basis for the set of relevant local operators (such as tr(φ
m
),
tr(∂
2
φ), etc...). γ
mn
is a “metric” on the space of operators, which we may take to be
a matrix of numerical constants by assuming that all masses and coupling constants are
included in the Os. Note that terms in the effective action at higher loops will involve
products of more than two operators, however the one loop terms may always be written
in this form. We now introduce a χ field coupling to each O,
S
χφ
=

d
d

n
(x)O
n
(φ(x)), (2.21)
and assume that the fields χ have propagators
χ
m
(p)χ
n
(−p) = −γ
mn
ln

p ◦ p +
1
Λ
2
p ◦ p

(2.22)
which could arise, for example, from the nonlocal quadratic action
S
χχ
= −

d
d
p
1
2
χ
m
(p)χ
n
(−p)γ
mn
¸
ln(
p ◦ p +
1
Λ
2
p ◦ p
)
−1
. (2.23)
Here, γ
mn
is the inverse
1
of γ
mn
.
1
If γ is not invertible, we have simply introduced too many χs. In this case, we choose a
new basis {
ˆ
O
n
} of operators such that some of the basis elements do not appear in (2.20). The
submatrix ˆ γ of γ corresponding to the
ˆ
Os which do appear will then be invertible, and we may
introduce χs as above coupling to this smaller set of operators. The kinetic term (2.23) for the
χs is then well defined, since we replace γ with ˆ γ.
8
2.5. Linear divergences in φ
4
in three dimensions
Before closing this section, we note that linear IR singularities at one loop may be
understood in a similar manner. For example, the commutative φ
4
theory in d = 3 has
a linear divergence in the two point function at one loop. In the noncommutative theory,
this leads to a term
S
IR
=

d
3
p
1
2
tr(φ(p))tr(φ(−p))
π
2
6(p ◦ p +
1
Λ
2
)
1
2
(2.24)
in the quadratic effective action which has a 1/|p| singularity for Λ = ∞. In order to
reproduce this singularity, we again introduce a χ field coupling to tr(φ) but this time we
need a propagator
χ(p)χ(−p) ∝
1
(p ◦ p)
1
2

1
(p ◦ p +
1
Λ
2
)
1
2
. (2.25)
3. What can give logarithmic and inverse linear propagators?
In the previous section we showed that the singular IR behavior of the one loop
effective actions for various scalar field theories can be reproduced by a Wilsonian action
which includes new particles χ
i
coupling linearly to various operators built from of φ,
assuming that the fields χ have logarithmic (or inverse linear) propagators. We would now
like to understand what dynamics could give rise to these propagators.
3.1. Spectral representation of propagators
As a first step, it is useful to rewrite the propagators in a spectral representation. We
introduce a parameter
2
α

with dimensions of squared length and a metric g
µν
= −

2
)
µν


)
2
then rewrite the propagator as
∆(p) =

dm
2
ρ(m
2
)
1
p◦p


)
2
+ m
2
. (3.1)
The logarithmic propagators
∆(p) = ln

p ◦ p +
1
Λ
2
p ◦ p

(3.2)
2
We call the constant α

since for noncommutative field theories arising from string theory,
the metric g
µν
= −

2
)
µν


)
2
is exactly the closed string metric in the zero slope limit of [21] (the
open string metric is G
µν
= δ
µν
) when α

has its usual meaning.
9
are reproduced with the spectral density
ρ(m
2
) =

1 m
2

1


Λ)
2
0 m
2
>
1


Λ)
2
(3.3)
suggesting a continuum of states with m
2
uniformly distributed between 0 and a cutoff
1/(α

Λ)
2
. Thus, we may replace χ with a continuum of states ψ
m
which have ordinary
propagators
ψ
m
(p)ψ
m
(−p) ∝
1
p◦p


)
2
+ m
2
(3.4)
which couple to φ (or some O(φ)) in a way that is independent of m,
S
ψφ
=
1


Λ)
2
dm
2

d
d

m
(x)φ(x). (3.5)
Note that for Λ →∞, the ψs completely decouple, leaving the original φ theory, as desired.
The ∆(p) =
1
(p◦p)
1
2
propagators, are reproduced by a spectral density
3
ρ(m
2
) =

1
mπα

m
2
<

4
(Λπα

)
2
0 m
2
>

4
(Λπα

)
2
.
(3.6)
As above, we may interpret this as a continuum of states ψ
m
coupling to φ, this time with
density 1/(α

m) up to a cutoff m
2

1


Λ)
2
.
3.2. χs as degrees of freedom in extra dimensions
A very simple possibility for the interpretation of the continuum of degrees of freedom
ψ
m
is that they are the transverse momentum modes of a particle ψ which propagates
freely in more dimensions. The continuous parameter m is related to the momentum q
in these new dimensions through m
2
= −q
2
. That is, we imagine that the d-dimensional
space in which the φ quanta propagate is a flat d-dimensional brane residing in a d + n
dimensional space. The ψ particles propagate freely in this space but couple to the φ
particles on the brane (located at x

= 0). We choose the metric seen by the ψ particles
to be g
µν
= −

2
)
µν


)
2
in the brane directions and δ
µν
in the transverse directions.
3
The symbols
>

and
<

are used because the second term in (2.25) is regularization dependent
and its precise functional form is not reproduced by choosing a sharp cutoff here, though the
behavior for p →0 at fixed Λ and for Λ →∞ is the same.
10
By choosing the number of extra dimensions to be one or two, we can precisely re-
produce the spectral densities (3.6) or (3.3) corresponding to inverse linear or logarithmic
propagators respectively.
Explicitly, with two extra dimensions we have
ψ(p, x

=0)ψ(−p, x

=0) =
1
α

Λ
d
2
q
(2π)
2
1
p◦p


)
2
+ q
2
=
1


Λ)
2
dm
2

1
p◦p


)
2
+ m
2
=
1

ln

p ◦ p +
1
Λ
2
p ◦ p

,
(3.7)
giving exactly the desired form of the logarithmic propagators. Note that we impose a
cutoff 1/(Λα

)
2
on q
2
to reproduce the cutoff in the spectral function (3.6). This cutoff on
transverse momenta decreases to zero as Λ →∞, but this provides the desired decoupling
of ψ in this limit.
With one extra dimension, we find
ψ(p, x

=0)ψ(−p, x

=0) =
2
Λπα

dq

1
p◦p


)
2
+ q
2
=
4
(Λπα

)
2
dm
2
1
2πm
1
p◦p


)
2
+ m
2
=
α

2(p ◦ p)
1
2

α

π(p ◦ p)
1
2
tan
−1

Λπ(p ◦ p)
1
2
2

(3.8)
thus giving a spectral density of 1/m and reproducing the correct p → 0 behavior of
the inverse linear propagators (2.25) above (as discussed in footnote 3, the difference in
functional form between the second terms here and in (2.25) is a consequence of the choice
of regularization scheme).
Actually, it is possible to see more directly that the theory with free ψ particles in two
extra dimensions coupling linearly to φs on the brane is precisely equivalent to a theory
with particles χ that live on the brane and have logarithmic propagators. Noting that it
is ψ(x, x

= 0) that couples to φ, we define χ(x) = ψ(x, x

= 0) and rewrite the ψ action
using a Lagrange multiplier λ(x) as
e

d
d
x φ(x)ψ(x,x

=0)−

d
d
xd
2
x

1
2
(∂ψ)
2
=

[dλ][dχ]e

d
d
x [φ(x)χ(x)+iλ(x)[χ(x)−ψ(x,x

=0)]]−

d
d
xd
2
x

1
2
(∂ψ)
2
=

[dλ][dχ]e

d
d
p [φ(p)χ(−p)+iλ(−p)χ(p)]−

d
d
pd
2
q [
1
2
ψ(−p,−q)(p◦p+q
2
)ψ(p,q)−iλ(−p)ψ(p,q))]
.
(3.9)
11
We may then integrate out ψ directly from this action, leaving
e

d
d
p

φ(p)χ(−p)+iλ(p)χ(−p)+
1
2
λ(p)λ(−p)

d
2
q
p◦p+q
2

.
(3.10)
Finally, integrating out the Lagrange multiplier λ gives
e

d
d
p

φ(p)χ(−p)+
1
2
χ(−p)χ(p) ln
−1

p◦p+
1
Λ
2
p◦p

(3.11)
as desired.
In the general case, we find that the nonlocal quadratic action (2.23) derived above
may be replaced by an ordinary, local higher dimensional kinetic term
ˆ
S
ψψ
= −
1

d
d+2
x
1
2
g
µν

µ
ψ
m

ν
ψ
n
γ
mn
, (3.12)
where g
µν
is taken to be −

2
)
µν


)
2
in the original d directions and δ
µν
in the transverse
directions.
As an example, the logarithmic terms in the φ
3
theory in six dimensions may be
reproduced using

d
6
x
gm
3 · 32π
φ(x)ψ(x, x

=0) +

3g
2
16πm
φ
2
(x) −
g
192πm

2
φ(x)

ˆ
ψ(x, x

=0)
+

d
8
x
1
2
g
µν

µ
ˆ
ψ∂
ν
ˆ
ψ −
1
12
g
µν

µ
ψ∂
ν
ˆ
ψ.
(3.13)
3.3. String theory analogy
The suggestion that certain high momentum degrees of freedom in φ are dual to fields
which propagate in extra dimensions fits rather nicely with the analogy that associates φs
with open strings and χs or ψs with closed strings. In noncommutative gauge theories that
arise as limits of string theory, the fields (φs) in terms of which the theory is defined are
modes of open strings living on a D-brane. In this case, there is a physical bulk in which
closed strings propagate, and the low energy closed string modes are related to high energy
modes of open strings by channel duality. In particular, a nonplanar one loop diagram of
the type we have studied is topologically equivalent to a string diagram in which a number
of open strings become a closed string which then turns back into open strings. The regime
in which the open string loop has very high momenta may be viewed equivalently as the
exchange of a very low momentum closed string, free to propagate in the bulk. This fits
very well with our interpretation that the IR singularities of nonplanar one loop diagrams
should be reproduced by tree level exchanges of a particle ψ (see figure 3). The connection
between ψ and closed strings is further strengthened by the fact that the ψs do not carry
any matrix indices, and as noted above, the ψs propagate in a metric which is precisely
the closed string metric identified in [21].
12
Fig. 3: Channel duality in string theory provides a natural understanding
of the correspondence between nonplanar one loop diagrams and tree level χ
exchange diagrams.
3.4. Interpretations of the extra dimensions
Despite the many similarities between the ψ fields and closed strings, the calculations
we have presented so far do not really indicate whether the extra dimensions in which the
ψs propagate are truly physical. There seem to be three logical possibilities:
1) The first possibility is that the “extra dimensions” are simply a mathematical conve-
nience - a simple way to state that χ has a logarithmic propagator.
2) At the other extreme is the possibility that the extra dimensions are real and the ψs are
propagating fields living in these dimensions whose restrictions to the d dimensional
space are the χ fields.
3) The third possibility is a compromise between these two: in situations where the
noncommutative field theory is a limit of string theory, the extra dimensions are
really the bulk dimensions transverse to the brane on which the φ fields live and the
extra dimensions are physical. Otherwise, they are only a mathematical convenience.
We note that in the case of noncommutative field theories arising from string theory, the
number of dimensions transverse to the brane is typically larger than two, so the propagator
of a single massless bulk field between points on the brane is nonsingular. However, in these
theories there is no reason to expect that only a single closed string mode ψ contributes.
Rather, the spectral function required to reproduce IR singularities would presumably arise
from the combination of extra dimensions and the density of closed string states which are
allowed to couple to the worldvolume field of interest.
4
4. Higher Loops
4.1. The three point function of φ
3
at two loops
In this section, we explore the infrared singularities appearing in Γ
(3)
at two loops
for the noncommutative φ
3
theory in six dimensions. In particular, we consider the two
4
We thank Lenny Susskind and Igor Klebanov for helpful discussions on this point.
13
p
3
p
2
k
k
k
k k
k
p p p
p
1
1
1
1
2
2
2
3
3
3
Fig. 4: The two contributions to the (trφ)
3
term in the effective action at
two loops.
diagrams of figure 4 which give the leading contribution to the (trφ)
3
term in the effective
action.
These diagrams have the property that each external line connects to a different index
loop in double line notation. The remaining two loop diagrams with three external lines
give subleading contributions either to tr(φ)tr(φ
2
) or to tr(φ
3
) terms.
The first diagram contributes a term to the effective action

d
6
p
1
d
6
p
2
d
6
p
3
δ(p
1
+ p
2
+ p
3
)φ(p
1
)φ(p
2
)φ(p
3
)V
1
(p
1
, p
2
, p
3
) , (4.1)
where
V
1
(p
1
, p
2
, p
3
) =
g
5
3 · 2
11
π
6
×

d
6
k
1
d
6
k
2
d
6
k
3
δ(k
1
+ k
2
+ k
3
)e
ik
1
×p
2
−ik
3
×p
3
(k
2
1
+ m
2
)((k
1
+ p
1
)
2
+ m
2
)(k
2
2
+ m
2
)((k
2
+ p
2
)
2
+ m
2
)(k
2
3
+ m
2
)((k
3
+ p
3
)
2
+ m
2
)
,
(4.2)
and k × p ≡ k
µ
θ
µν
p
ν
. The external momenta appearing in the denominator do not con-
tribute to the infrared singular terms, since by Taylor expanding the denominators in
momenta, we find that the subleading terms are nonsingular for p → 0. To evaluate the
leading terms, we rewrite the delta function as

e
iy·(k
1
+k
2
+k
3
)
to obtain
V
1
(p
1
, p
2
, p
3
) =
g
5
3 · 2
11
π
6

d
6
y
(2π)
6
3
¸
i=1

d
6
k
i
e
ik
i
·x
i
(k
2
i
+ m
2
)
2
, (4.3)
where x
1
= y + θp
2
, x
2
= y, x
3
= y −θp
3
. The IR singular terms come from the region of
small y, so we need the behavior of the k integrals for small x.
In this regime, we have

d
6
k
i
e
ik
i
·x
i
(k
2
+ m
2
)
2
=

3
x
2
−m
2
π
3
ln(
1
x
2
m
2
) + . . . . (4.4)
14
Only the leading term contributes to the IR singular pieces, so our Feynman integral
becomes
V
1
(p
1
, p
2
, p
3
) =
g
5
π
3
3 · 2
5

Λ
d
6
y
(2π)
6
1
y
2
(y + θp
2
)
2
(y −θp
3
)
2
. (4.5)
The cutoff Λ has been included because our approximations are valid only for small y,
but this is the only part of the integral that gives the IR singular terms of interest. This
integral clearly has a logarithmic singularity as all momenta go to zero, but is finite, if
any one momentum is scaled to zero. By a careful analysis of the integral (4.5), it may be
shown that for small momenta, we have
V
1
(p
1
, p
2
, p
3
) =
g
5
3 · 2
12
ln

1
p
1
◦ p
1
+ p
2
◦ p
2
+ p
3
◦ p
3

+ regular terms. (4.6)
We now turn to the second diagram of figure 3. This contributes a term to the effective
action

d
6
p
1
d
6
p
2
d
6
p
3
δ(p
1
+ p
2
+ p
3
)φ(p
1
)φ(p
2
)φ(p
3
)V
2
(p
1
, p
2
, p
3
) , (4.7)
where
V
2
(p
1
, p
2
, p
3
) =
g
5
2
10
π
6

d
6
k
1
d
6
k
2
d
6
k
3
δ(k
1
+ k
2
+ k
3
)e
ik
3
×p
2
−ik
1
×p
3
(k
2
3
+ m
2
)((k
3
+ p
2
)
2
+ m
2
)(k
2
1
+ m
2
)((k
1
+ p
3
)
2
+ m
2
)
·
1
((k
1
+ p
3
+ p
1
)
2
+ m
2
)(k
2
2
+ m
2
)
.
(4.8)
As above, the momenta in the denominator do not affect the IR singular terms and we
may rewrite the integral as
V
2
(p
1
, p
2
, p
3
) =
g
5
2
10
π
6

d
6
y
(2π)
6

d
6
k
1
d
6
k
2
d
6
k
3
e
ik
i
·x
i
(k
2
1
+ m
2
)
3
(k
2
2
+ m
2
)(k
2
3
+ m
2
)
2
, (4.9)
with x
1
= y −θp
3
, x
2
= y, x
3
= y +θp
2
. The IR singular terms come only from the region
of small y. For small x, we have

d
6
k
i
e
ik
i
·x
i
(k
2
+ m
2
)
=
16π
3
x
4


3
m
2
x
2
+
1
2
m
4
π
3
ln(
1
x
2
m
2
) + . . .

d
6
k
i
e
ik
i
·x
i
(k
2
+ m
2
)
2
=

3
x
2
−m
2
π
3
ln(
1
x
2
m
2
) + . . .

d
6
k
i
e
ik
i
·x
i
(k
2
+ m
2
)
4
=
1
2
π
3
ln(
1
x
2
m
2
) + . . . .
(4.10)
15
The IR singular terms come from taking the leading term in these three expressions, and
we find
V
2
(p
1
, p
2
, p
3
) =
g
5
π
3
2
5

Λ
d
6
y
(2π)
6
1
y
4
(y + θp
2
)
2
ln

1
(y −θp
3
)
2

. (4.11)
The small momentum behavior of this integral when any one or all three momenta are
scaled to zero is reproduced by the function
ˆ
V
2
(p
1
, p
2
, p
3
) =
g
5
2
11
(
1
2
ln(p
2
◦p
2
) ln(p
2
◦p
2
+p
3
◦p
3
)−
1
4
ln
2
(p
2
◦p
2
+p
3
◦p
3
)−
1
4
ln(p
2
◦p
2
+p
3
◦p
3
)) .
(4.12)
Note that the logarithmic divergence that arises as p
2
→ 0 comes from the three-point
subdiagram which would be divergent in the commutative theory. On the other hand,
setting p
1
or p
3
to zero does not lead to a divergence. It is possible that there is some
other function with the same singularities as (4.12) whose form would be more suggestive
of an interpretation by χs.
For φ
3
theory with a single φ, we should sum the contribution of V
1
with that of V
2
(which is automatically symmetrized in momenta when inserted into the expression (4.7))
to get the complete two loop contribution to the (trφ)
3
term in the effective action. In
writing a Wilsonian action to reproduce the IR singularities of the φ
3
theory, we do not
necessarily need to match the behavior diagram by diagram; however it is possible to take
a theory which isolates the contribution V
1
, for example. In the theory with Lagrangian
density
tr

1
2
¸
i
(∂φ
i
)
2
+
1
2
m
2
¸
i
φ
2
i
+ g(φ
1
⋆ φ
4
⋆ φ
4
+ φ
2
⋆ φ
5
⋆ φ
5
+ φ
3
⋆ φ
6
⋆ φ
6
+ φ
4
⋆ φ
5
⋆ φ
6
+ c.c.)

,
(4.13)
the leading tr(φ
1
)tr(φ
2
)tr(φ
3
) term in the effective action is precisely V
1
. Thus, we would
like to understand how the singularity ln(p
2
1
+ p
2
2
+ p
2
3
) alone or terms of the form (4.12)
could arise from a Wilsonian effective action.
4.2. Interpretation of the singularities
The two loop IR singularities we have found have momentum dependence that cannot
be reproduced by tree level diagrams with local couplings only on the brane (such as
the tree level χ exchange diagrams which reproduced all one loop singularities). Each
propagator and vertex factor of such a diagram is a function of a single sum of external
16
a b
Fig. 5: Possibilities for the interpretation of two loop IR singularities.
momenta, while the singularities we have found, (4.6) and (4.12), cannot be reproduced
by a product of such functions.
If our interpretation of the ψ fields as higher dimensional particles is correct, one
possibility would be that these singularities arise from a diagram like that of Figure 5a,
in which each φ becomes a ψ and these three ψ field interact locally in the bulk. This is
suggested both by the (trφ)
3
structure and by the fact that viewed as worldsheet diagrams
in string theory, the double line versions of the diagrams in figure 4 are topologically
equivalent to the closed string interaction diagram in figure 6.
Fig. 6: Closed string interaction diagram topologically equivalent to the
diagrams of figure 4.
There are various possibilities for the form of a ψ
3
interaction, including possible
derivatives on the ψs and the possibility of a coupling that varies as some function of the
transverse coordinates x

(for example, a varying dilaton). Though such interactions do
seem to give IR singularities with nontrivial dependence on external momenta, we have
not found a simple action with local interactions of the ψs in the bulk that reproduces the
two loop singularities above.
A second possibility would be that the momentum dependence of the two loop IR
singularities arises from a loop of χs (or ψs), such as the diagram of figure 5b. The form
of the expressions (4.5) and (4.11) are suggestive of such a diagram, if we reinterpret the
17
y integrals as integrals over a loop momentum l = θy. For example, the expression (4.5)
becomes
V
1
(p
1
, p
2
, p
3
) =
g
5
π
3
3 · 2
5
det(Θ)

Λ
d
6
l
(2π)
6
1
l ◦ l (l + p
2
) ◦ (l + p
2
) (l −p
3
) ◦ (l −p
3
)
. (4.14)
which is exactly reproduce by the diagram of figure 5b with a coupling proportional to
g
5
3
det
1
3
(Θ)φ(x)χ
0
(x)χ
0
(x) where χ
0
propagates on the brane with metric g
µν
. Such a
coupling is undesirable, however, since it would lead to higher point functions with frac-
tional powers of the coupling and more severe IR singularities which are not present in the
original theory.
To conclude, we do not have a satisfactory interpretation of the two loop IR singu-
larities in terms of weakly coupled light degrees of freedom. We hope to return to this
important problem in the near future.
Acknowledgements
We would like to thank T. Banks, D. Gross, I. Klebanov, S. Minwalla, G. Semenoff,
L. Susskind, and E. Witten for useful discussions. The work of M.V.R. was supported in
part by NSF grant PHY98-02484 and that of N.S. by DOE grant #DE-FG02-90ER40542.
18
References
[1] T. Filk, “Divergences in a Field Theory on Quantum Space,” Phys. Lett. B376 (1996)
53.
[2] J.C. Varilly and J.M. Gracia-Bondia, “On the ultraviolet behavior of quantum
fields over noncommutative manifolds,” Int. J. Mod. Phys. A14 (1999) 1305, hep-
th/9804001.
[3] M. Chaichian, A. Demichev and P. Presnajder, “Quantum Field Theory on Non-
commutative Space-times and the Persistence of Ultraviolet Divergences,” hep-
th/9812180; “Quantum Field Theory on the Noncommutative Plane with E(q)(2)
Symmetry,” hep-th/9904132.
[4] M. Sheikh-Jabbari, “One Loop Renormalizability of Supersymmetric Yang-Mills The-
ories on Noncommutative Torus,” hep-th/9903107, JHEP 06 (1999) 015.
[5] C.P.Martin, D. Sanchez-Ruiz, “The One-loop UV Divergent Structure of U(1) Yang-
Mills Theory on Noncommutative R
4
,”hep-th/9903077, Phys.Rev.Lett. 83 (1999) 476-
479.
[6] T. Krajewski, R. Wulkenhaar, “Perturbative quantum gauge fields on the noncommu-
tative torus,” hep-th/9903187.
[7] S. Cho, R. Hinterding, J. Madore and H. Steinacker, “Finite Field Theory on Non-
commutative Geometries,” hep-th/9903239.
[8] E. Hawkins, “Noncommutative Regularization for the Practical Man,” hep-th/9908052.
[9] D. Bigatti and L. Susskind, “Magnetic fields, branes and noncommutative geometry,”
hep-th/9908056.
[10] N. Ishibashi, S. Iso, H. Kawai and Y. Kitazawa, “Wilson Loops in Noncommutative
Yang-Mills,” hep-th/9910004.
[11] I. Chepelev and R. Roiban, “Renormalization of Quantum Field Theories on Non-
commutative R
d
, I. Scalars,” hep-th/9911098.
[12] H. Benaoum, “Perturbative BF-Yang-Mills theory on noncommutative R
4
,” hep-
th/9912036.
[13] S. Minwalla, M. Van Raamsdonk, N. Seiberg “Noncommutative Perturbative Dynam-
ics,” JHEP 02 (2000) 020, hep-th/9912072.
[14] G. Arcioni and M. A. Vazquez-Mozo, “Thermal effects in perturbative noncommuta-
tive gauge theories,” JHEP 0001 (2000) 028, hep-th/9912140.
[15] M. Hayakawa, “Perturbative analysis on infrared aspects of noncommutative QED on
R**4,” hep-th/9912094; “Perturbative analysis on infrared and ultraviolet aspects of
noncommutative QED on R**4,” hep-th/9912167.
[16] S. Iso, H. Kawai and Y. Kitazawa, “Bi-local fields in noncommutative field theory,”
hep-th/0001027.
19
[17] H. Grosse, T. Krajewski and R. Wulkenhaar, “Renormalization of noncommutative
Yang-Mills theories: A simple example,” hep-th/0001182.
[18] I. Y. Aref’eva, D. M. Belov and A. S. Koshelev, “A note on UV/IR for noncommutative
complex scalar field,” hep-th/0001215.
[19] W. Fischler, E. Gorbatov, A. Kashani-Poor, S. Paban, P. Pouliot and J. Gomis, “Ev-
idence for winding states in noncommutative quantum field theory,” hep-th/0002067.
[20] A. Matusis, L. Susskind and N. Toumbas, “The IR/UV connection in the non-
commutative gauge theories,” hep-th/0002075.
[21] N. Seiberg and E. Witten, “String Theory and Noncommutative Geometry,” hep-
th/9908142, JHEP 09 (1999) 032.
20

Sign up to vote on this title
UsefulNot useful